New Data on Hill-forts of North-Western Latvia

Transcription

1 ISSN There are some 470 hill-forts in Latvia, 80 of which have seen large- or small-scale excavation (Vasks, 2005, p ). The majority of these 80 hill-forts are located in eastern Latvia, and only nine have been excavated in the western part of the country. Out of these nine, only Talsi Hill-Fort has been excavated on a large scale, while the rest have been subject only to limited excavation. Accordingly, archaeological research in western Latvia, including hill-fort research, has been given greater attention in recent years. Since 2002 the Department of Archaeology and Ancillary Historical Sciences of the University of Latvia has organised the excavation of several archaeological sites in western Latvia in the frame of student field courses. The excavated sites include three hill-forts: Beltes, Mežīte and Puze (Fig. 1). Beltes Hill-Fort The most extensively excavated of these, Beltes (Padures) Hill-Fort, is located on the left bank of the River Venta. It was established on a promontory separated on the north and south sides by deep ravines, and on the east side by the steep bank of the Venta. On the open western side a ditch had been dug, and a 4-mhigh bank thrown up on the plateau. To the west and south there was an adjoining settlement site covering about 3 ha (Fig. 2). In terms of its topographic location and the character of the system of defences, Beltes Hill-Fort belongs to the type of hill-fort most common in western Latvia. Excavation was undertaken at Beltes Hill-Fort in 2003, 2005, 2006 and 2007 (Asaris & Vasks, 2004, p ; Vasks, 2006 a, p ; Vasks, 2008, p ). During this period an area of 279 m² was excavated on the hill-fort, out of a total area of about 1900 m² (14.7 % of the plateau). On removal of the cultural layer, m thick and up to m in places, a series of features relating to the structures on the hill- fort were uncovered (hearths, stoves, various pits, postholes, etc.). In the course of the excavation 624 objects were recovered, as well as 8460 pot-sherds. Bones of mammals, birds and fish were collected as well. Beltes Hill-Fort was inhabited in two phases. The early phase included the Late Bronze and Pre-Roman Iron Age, while the late phase included the second half of the Middle Iron Age and the Late Iron Age. The fort and the settlement that had grown up around it went out of existence in the second half of the 12 th or the early 13 th century. Pollen analysis of the deposits indicates two periods of human economic activity. The first begins in the final part of the Atlantic (AT3), BP. In terms of archaeological periodisation, this corresponds to the Middle Neolithic. During this period there was alder-birch forest growing in the environs of the hillfort, with a significant amount of wych elm, lime and hazel. At the beginning of the Subboreal (Middle Neolithic) the proportion of spruce increased, and agriculture evidently began in the environs of the hillfort (oats and hemp pollen in the deposits), resulting in the development of a partially open landscape with meadows and pasture. The second period of greater activity occurs in the Subatlantic, which begins at 2800 BP, corresponding to the Bronze and Iron Age (SA1, SA2), namely the time when Beltes Hill-Fort developed. At this time the hillfort was no longer surrounded by forest and, judging from the rapid rise in pollen of ruderals, meadow and pasture plants, and various herbs, compared with the earlier period of activity, there was major agricultural activity in the environs of the hill-fort. Rye is also recorded, in addition to oats and hemp (Vasks, Kalniņa, Daugnora, 2011). Radiocarbon dates indicate that the fortified habitation developed on the headland at the bank of the River Venta, delimited by ravines, at the end of the 2 nd mill. BC, when a wooden palisade was erected (

2 Fig. 1. The hill-forts of western Latvia. 1 pav. Vakarų Latvijos piliakalniai 930 cal. BC) (Bērziņš et al., 2009). Beltes Hill-Fort is so far the only one in western Latvia where Late Bronze Age occupation is confirmed by radiocarbon dating. This also indicates that the earliest hill-forts developed in western Latvia approximately at the same time as in eastern Latvia and eastern Lithuania (Grigalavičienė, 1995, p. 27). The area of the hill-fort delimited by the palisade (Fig. 3) constituted about half of the area that the hill-fort occupied during the Late Iron Age. However, it seems that already in the early phase of occupation, as the population rapidly increased, the inhabited area of the plateau was extended, moving the palisade further to the south-west. The data on the structures of this phase is scanty: we can only say that they were above-ground post-built structures with internal stone-lined hearths. 65

3 I VIII I II III Fig. 2. Plan of Beltes Hill-Fort showing areas excavated in (drawing: E. Brastiņš, N. Grasis, A. Vasks). I excavation areas. II extent of the settlement (early town) next to the hill-fort. III location of the approach to the hill-fort. 2 pav metais vykdytų kasinėjimų Beltes piliakalnyje planas (braižė: E. Brastiņš, N. Grasis, A. Vasks). I kasinėtas plotas. II gyvenvietės (ankstyvojo miesto), esančios šalia piliakalnio, dydis. III piliakalnio prieigos An interesting feature was discovered when the cultural layer was removed. Two stones were found at the level of the natural subsoil, one of which, fairly large (60 80 cm), had a flat surface, while the one next to it was smaller and elongated, with a sloping polished surface at one end (Fig. 4). The significance of this feature was quite clear: this was a place or workshop for grinding stone and bone tools. As the members of the excavation team could ascertain for themselves, the person engaged in this work could sit comfortably on the large, flat stone, holding the object to be worked in both hands and grinding it against the end of the elongated stone, which was firm in the ground. Judging from the presence nearby of three or four post-holes, the workshop had been located within a shelter or even a building of some kind. The oats and hemp identified in the pollen indicates cultivation. Although the presence of two-row barley and emmer wheat is not definitely confirmed, since the identification is simply of cultivated cereals as such (Cerealia), nevertheless the idea that these two cereals in particular were cultivated at Beltes Hill-Fort should not be abandoned, considering that both are characteristic of many Late Bronze Age hill-forts of the East Baltic (Rasiņš & Tauriņa, 1983, p. 152, 163). Further palynological research is required. Almost 78 % of the faunal remains were from domestic stock (Table 1). This figure corresponds ap- 66

4 Fig. 3. Beltes Hill-Fort. Excavation Area III, with uncovered belt of posts of palisade (photo A. Vasks). 3 pav. Beltes piliakalnis. Kasinėtas III plotas, kuriame atidengta stulpaviečių eilė (A. Vasks nuotrauka) Fig. 4. Beltes Hill-Fort. A workshop for grinding stone and bone tools. 4 pav. Beltes piliakalnis. Akmeninių ir kaulinių įrankių gamybos dirbtuvė 67

5 Table 1. Bone fragment counts of animal species, early phase of occupation* 1 lentelė. Gyvūnų rūšių kaulų fragmentų kiekis, ankstyvasis apgyvendinimo laikotarpis Cattle Sheep/Goat Pig Horse Wild animals Aves Fishes Bos taurus Ovis aries/capra hircus Sus sforca Equus caballus % 9.52 % % % % 3.17 % 2.65 % % % Table 2. Bone fragment counts of animal species, late phase of occupation* 2 lentelė. Gyvūnų rūšių kaulų fragmentų kiekis, vėlyvasis apgyvendinimo laikotarpis Cattle Sheep/Goat Pig Horse Wild animals Aves Fishes Bos taurus Ovis aries/capra hircus Sus sforca Equus caballus % % % 6.44 % % 3.27 % % % % * The bone material was determined by Dr. L. Daugnora Gyvūnų kaulus identifikavo dr. L. Daugnora proximately to what has been found at other hill-forts in the East Baltic (Vasks, 1994, Table 7). Judging from the bone count, horses were consumed the most. These are followed by pig and cattle and then by sheep/goats. Such a ranking of domestic stock in terms of their importance for subsistence is somewhat unusual, since cattle were generally the most common at Late Iron Age hill-forts, with pigs and sheep/goats sharing second and third place, and horses coming after that (cf. Graudonis, 1989, Table 9). However, it has to be considered that the number of domestic animal bones from the first occupation phase is small (147 fragments), so the proportions of the different species could be fortuitous. Among wild animals, aurochs was hunted the most: aurochs bones constituted half of all wild animal bones (Vasks, Kalniņa, Daugnora, 2011). Cultural affiliation in the early occupation phase is indicated by the character of the pottery. Striated pottery predominates, and so the hill-fort can be included among the sites with this kind of pottery (Table 3). As at other sites in western Latvia from this time, right from the beginning of occupation, early rusticated pottery was present at Beltes Hill-Fort. This is the first hill-fort in western Latvia where textile-impressed pottery has been found (one more site where such pottery later has been found is open settlement of Priednieki; see Vasks, 2006 a, p. 66). The hill-fort has not produced any evidence of habitation in the Roman Iron Age or the beginning of the Middle Iron Age, so it may be presumed that the hill-fort was not occupied at this time. Habitation recommenced in the second half of the Middle Iron Age, the 6 th 7 th century, as indicated by some finds of artefacts characteristic of this period. The most activity occurred at the hill-fort in the 9 th 12 th century, and it is possible that the major defensive bank, the entrance terrace along the outer slope of the bank, and the ditch were created at this time. At this time a settlement with an area of about 3 ha formed next to the hill-fort. During this late period of occupation the hill-fort plateau was densely built-up, mainly with above-ground buildings (probably log structures), although there were sunken buildings as well. In some cases round pits used for household purposes were dug beneath the above-ground buildings. These measured about 1 m in diameter and had a flat bottom. Heating was provided by stoves both stoves of piled stones and clay stoves and to a lesser extent by stone-lined hearths. The economy is characterised by the finds of tools: knives, awls, shears, warp weights, a spindle whorl, a scythe fragment, crucible fragments, etc. The late phase of occupation exhibits two characteristic that differ from those observed in eastern Latvia. In the first place, pottery has not been used as much, something that might be explained in terms of the wider use of wooden and bark vessels. This, in turn, could be the 68

6 Table 3. Pottery wares from Areas I and III, Beltes Hill-Fort (early phase of occupation) (number of sherds and %) 3 lentelė. Keramika iš Beltes piliakalnio I ir III plotų (ankstyvasis apgyvendinimo laikotarpis) (šukių skaičius ir %) Smooth Striated Rusticated Early rusticated Polished burnished Pinched Textileimpressed Wheel-made Total ,7 % ,1 % 111 4,0 % 226 8,1 % 13 0,5 % 25 0,9 % 55 2,0 % 20 0,7 % % Table 4. Pottery wares from Beltes Hill-Fort, Areas II, VII and VIII (late phase of occupation) (number of sherds and %) 4 lentelė. Keramika iš Beltes piliakalnio II, VII ir VIII plotų (vėlyvasis apgyvendinimo laikotarpis) (šukių skaičius ir %) Smooth Striated Rusticated Early rusticated Polished burnished Pinched Textileimpressed Wheel-made Total ,8 % ,6 % ,5 % 9 0,6 % 51 3,6 % 13 0,9 % 2 0,1 % 55 3,9 % % reason for the second characteristic, namely the late introduction of wheel-made pottery into the everyday life of the people at Beltes Hill-Fort and the settlement site (Table 4). Agriculture, about which scanty data has been obtained (and it is thought that rye appeared at this particular time), as well as stock-keeping provided the basis of subsistence. Domestic stock constituted 70 % of all the bones from the late phase of occupation (Table 2). Judging from the number of bones, beef was used most for food (evidently along with milk products), followed by sheep/goat and pigs. Compared with the early occupation phase, the consumption of horse meat had fallen considerably. This can be explained in terms of the increased role of these animals in various kinds of labour, and possibly in war as well. Chickens were also kept. A proportion of subsistence was also provided by wild animals: aurochs, elk, red deer and beaver. The last, along with the fox, was also hunted for its fur (Vasks, Kalniņa, Daugnora, 2011). Western European coins found on the hill-fort (and an 11 th century German coin found in the settlement) indicate trading contacts. True, of the five coins found on the hill-fort, three were imitations worn as pendants. Artefact chronology and radiocarbon dates indicate the time when the late phase of occupation ended, and thus the centre as such ceased to exist. The majority of artefacts date from the Late Iron Age, i.e. the 9 th 12 th century, and especially the 11th and first half of the 12 th century. There are no finds that can definitely be dated to the second half of the 12 th or the 13 th century. However, it should be taken into account that some of the finds have a fairly broad chronological range, the 11 th to 13 th century, so it is not possible to date the end of the late phase of occupation on the basis of the artefact dates. Eight radiocarbon dates fall in time interval between 820 and 1220 cal. AD, and three of them indicate the period between 1100 and 1220, i.e. the 12 th and early 13 th century (Bērziņš et al., 2009). Taking this into consideration, the time when Beltes Hill- Fort went out of existence as a centre can be dated to the second half of the 12 th or the early 13 th century. Indirect evidence that the hill-fort was no longer in existence in the 1230s is provided by the agreement of 1230 between the Couronians of the lower Venta area and papal legate Baldwin of Aulnes, and by the 1253 agreement between the Teutonic Order and the Bishop of Courland on the division of six Couronian lands. These two documents mention a whole series of places, some of which can be identified as located in the environs of Beltes Hill-Fort: Ķimale, Tigve and Lake Nabe. At the same time these documents do not include place names that could definitely be related to Beltes Hill-Fort and settlement. Evidently, this centre was no longer in existence in the first half of the 13 th century, when the agreements were concluded. Who were the inhabitants of Beltes Hill-Fort during the late phase of occupation? To some degree this can be determined by considering the kinds of orna- 69

7 Fig. 5. Beltes Hill-Fort. Ornaments and utensils from the late phase of occupation (drawing B. Vaska). 1, 2, 4 penannular brooches, 3 chain divider, 5 pendant ornament, 6 disc brooch, 7 tinkler with bead, 8, 15, 16 beads, 9 head of a dress-pin (?), 10 fragment of a cruciform pin, 11 mount of a knife sheath, 12, 19 key fragments, 13, 20, 23, 28 zoomorphic pendants, 14 tusk pendant, 17 mount, 18 spiral ring, 21 whetstone-casting mould, 22 single-piece double comb, 24 roarer, 25 fragment of the foot of a crossbow fibula, 26, 27 chain fragments. 1, 3, 4, 6, 7, 10, 11, 13, 18, 20, 23, bronze, 2, 9, 12, 19 iron, 5, 8, 15, 16 glass, 21 sandstone, 22, 24 bone. 5 pav. Beltes piliakalnis. Papuošalai ir kiti dirbiniai iš vėlyvosios apgyvendinimo fazės (piešinys: B. Vaska). 1, 2, 4 pasaginės segės, 3 skirstiklis, 5 kabutis, 6 plokštelinė segė, 7 karolis su žvangaliuku, 8, 15, 16 karoliai, 9 smeigtuko galvutė (?), 10 kryžinio smeigtuko fragmentas, 11 peilio makšties apkalas, 12, 19 rakto fragmentai, 13, 20, 23, 28 zoomorfiniai kabučiai, 14 ilties kabutis, 17 makštis, 18 įvijinis žiedas, 21 smiltainio liejimo formelė, 22 dvigubos šukos, 24 kaulo dirbinys, 25 lankinės segės kojelės fragmentas, 26, 27 grandinėlės fragmentai. 1, 3, 4, 6, 7, 10, 11, 13, 18, 20, 23, žalvaris, 2, 9, 12, 19 geležis, 5, 8, 15, 16 stiklas, 21 smiltainis, 22, 24 kaulas 70

8 Fig. 6. Mežīte Hill-Fort. A cellar pit beneath the above-ground structure (photo A. Vasks). 6 pav. Mežīte piliakalnis. Duobė po antžemine struktūra (nuotrauka: A. Vasks) ments found on the hill-fort (Fig. 5). Most of them have analogies among the material of northern Kurzeme, Saaremaa, the Livs of Vidzeme and the Vends, i.e. in the Finnic cultures in general. In view of these similarities, the inhabitants of the hill-fort during the Late, and possibly also the Middle Iron Age can be identified as belonging to one of the Finnic groups, possibly to the Vends mentioned in the Chronicle of Henry of Livonia. As indicated by the research undertaken to date, the Couronian expansion, pushing into northern Kurzeme, began in the 11th century and continued in the 12th. This can be seen very clearly from the spread of characteristic Couronian flat cremation cemeteries in northern Kurzeme, in the area where in the 11th 12th century inhumation had generally been practiced. Quite possibly, Beltes Hill-Fort and settlement site, the material culture of which does not show the influence of Couronian culture, was destroyed in the second half of the 12th century by the incoming Couronians. Then a new centre developed at Veckuldīga on the bank of the Venta, 6 km upstream from Beltes Hill-Fort. Mežīte Hill-Fort Mežīte is a complex of archaeological sites centring on a hill-fort, which is surrounded by a settlement with an area of about 4 ha. The complex of sites also includes three cemeteries and a cult site Elkakalns. The hill-fort is located on an isolated hill of glacial till: the slopes were steepened and a 2.5-m-high bank thrown up along the southern margin of the plateau. Although only 24 m2 of the total area of 2500 m2 of the hill-fort plateau have been excavated in the course of trial excavation in 2008 and 2009, the material recovered was very interesting. In the cultural layer, up to 2 m thick, the location of an above-ground structure was found, with the remains of an oven of piled stones. Beneath part of this building there was a cellar pit measuring 2 3 m, with a depth of 80 cm. The sides of the pit were quite steep and were lined with stones. The stratigraphy of the layers in the cellar, viewed in cross-section, indicates that the wooden covering of the cellar, covered with grey clay, had been renewed at least once (Fig. 6). Even though only a small area was 71

9 Fig. 7. Mežīte Hill-Fort. Ornaments, tools and weapons (drawing B. Vaska). 1, 2 crossbow bolt heads, 3 fish-hook, 4, 5, 6, 8, 10, 14, 16 pendant ornaments, 9 buckle of a belt, 7, 15 fingering, 11 three-ply bead, 12 mechanism of collapsible scales, 13 penannular brooche, 17 thinker, 18, 19 lenticular armbands, 20 knife, 21 ploughshare, 22 part of curb, 23 spearhead, 24, 25 awls. 1 3, iron, 4 10, 12 16, 18, 19 bronze, 11 glass, 17 clay. 7 pav. Mežīte piliakalnis. Papuošalai, įrankiai ir ginklai (piešinys: B. Vaska). 1, 2 arbaleto strėlių antgaliai, 3 žvejybos kabliukas, 4, 5, 6, 8, 10, 14, 16 kabučiai, 9 diržo sagtis, 7, 15 žiedai, 11 trigubas karolis, 12 sudedamųjų svarstyklių mechanizmas, 13 pasaginė segė, 17 kalavijo rankenos viršus, 18, 19 juostinės apyrankės, 20 peilis, 21 noragas, 22 žąslų dalis, 23 ietigalis, 24, 25 ylos. 1 3, geležis, 4 10, 12 16,18, 19 žalvaris, 11 stiklas, 17 molis 72

10 Fig.8. Puze Hill-Fort. Location of two burned buildings (photo: A. Vasks) 8 pav. Puze piliakalnis. Dviejų sudegusių pastatų vietos (nuotrauka: A. Vasks) excavated, a rich and diverse artefact assemblage was recovered, consisting of bronze ornaments, weapons and tools (Fig. 7). There are a string of finds indicative of trading activities. In the first place, there were seven coins. These include four Anglo-Saxon coins (Aethelred and Cnut) from the first half of the 11 th century, as well as 13 th century coins. The finds of the mechanism of collapsible scales (Fig. 7:12), as well as weights and a hacked-off piece of a forged silver bar indicate that trading was active. It should be noted that the people living in the settlement also engaged in trading activities: four weights and a coin of Cnut, from the first half of the 11 th century, were found in Area I (Guščika & Vasks, 2010, p ) Although the faunal remains from the hill-fort have not been precisely determined yet, it is unusual that in some samples fish bones constituted % of the total number of bones. Such a predominance of fish is not observed in the material from the settlement site. Nowadays there are no natural open water-bodies in Mežīte, but there are a number of bogs in the immediate vicinity of the hill-fort. Evidently, a thousand years ago there were lakes rich in fish. The presence of the cult site of Elkukalns, the three cemeteries and the extensive settlement site surrounding the hill-fort on all sides, with a cultural layer up to 1.5 m thick, all indicate that in the 11 th 13 th century Mežīte was an important centre for this area, with well-developed trade. What caused the abandonment of this centre is evident from the finds of five crossbow bolt heads (more of which were found in the settlement site; Fig. 7:1, 2). Evidently the hill-fort and settlement were destroyed in a crusader attack in the 13 th century. It is possible that this is the location of the central site of castellatura Lodgie, mentioned in a document from 1234, and the population centre Ladze, recorded in a document of Puze Hill-Fort Puze Hill-Fort differs from the two above-described hill-forts in terms of its small dimensions: its plateau measures only 600 m 2, of which 34 m 2 have been excavated in Another difference is the absence of a settlement site next to Puze Hill-Fort. On the plateau the cultural layer was 0.3 m thick. The locations of two buildings, extending down into the subsoil by up to 30 cm, were discovered (Fig. 8). One of these was located by the slope of the hill and is regarded as belonging to structures built along the perimeter of the plateau, or to the outer wall of the defences. In the central part of this log structure, measuring 3.5 x 4 m, was a stonelined hearth. On the site of the building and also outside it a large quantity of burned grain was recovered (soft wheat and rye, with a smaller quantity of barley; also field beans, peas, a small quantity of oats, as well as weeds among the charred plant remains). The location of the other building on the plateau was marked 73

11 Fig. 9. Puze Hill-Fort. A cache of bronze ornaments in situ (photo: A. Vasks). 9 pav. Puze piliakalnis. Žalvarinių papuošalų sankaupa in situ (nuotrauka: A. Vasks) by a pit in the natural subsoil, with a width of 1.6 m and a length of 2 m within the excavated area. It is possible that this was a cellar, similar to those discovered on Mežīte Hill-Fort, and that the building itself was larger. The iron objects recovered include a cylindrical key, an awl with a twisted mid-part, rivets and knives. Bronze artefacts include fragments of ribbonlike bracelets and fragments of fittings, as well as a finger-ring with a twisted head. The dating of the finds indicates that the hillfort was inhabited in the 12 th 13 th century. The majority of artefacts were strongly burned, and some had been transformed into amorphous lumps of molten bronze. This suggests that the fortification was destroyed in a major fire. A cache of bronze ornaments, constituting a set of women s jewellery, was also found in this structure (Fig. 9). It consisted of a neck-ring, chains with pendants, penannular brooches and an annular brooch, as well as finger-rings (Vijups, 2003). Strangely enough, Couronian ribbonlike bracelets, a very common form of jewellery at this time, were not present among the ornaments. It is possible that such bracelets were worn on an everyday basis, 6 12 on each arm, and consequently there were not removed and placed in the jewellery chest that we found in Puze. This jewellery can be interpreted as a stock of additional ornaments, kept in the storehouse and used to augment the set of jewellery on festive occasions. A sample of charred grain from the burned layer indicated the time of destruction of the defences: between 1220 and 1275 cal. AD, i.e. around the middle of the 13 th century (Bērziņš et al., 2009). This hillfort, too, was most probably destroyed in the course of the crusades. Conclusions This research on hill-forts, even though it has so far been undertaken on a small-scale, has generated several new ideas concerning the prehistory of western Latvia, which has still not been extensively researched. 1. As indicated by the radiocarbon dates for Beltes Hill-Fort, fortified settlement sites or hill-forts developed in western Latvia at the beginning of the Late Bronze Age, i.e. at the close of the 2 nd millennium BC, approximately at the same time as they appeared in eastern Latvia and Lithuania. The finds from Beltes Hill-Fort and the open settlement of Priednieki indicate that north-western Latvia can also be included in the area of distribution of textile-impressed pottery. 2. In the Late Iron Age Beltes Hill-Fort, and especially Mežīte Hill-Fort can be described as major centres, with large settlement sites, or early towns, next to them. If we compare the evidence of trade from Beltes Hill-Fort, located by an important waterway the 74

12 River Venta, with that obtained from the hill-forts of Talsi and Mežīte, where there is no such waterway, then we must admit that more evidence of this kind has been obtained from the other two hill-forts. At Talsi Hill-Fort, 13 Western European, Arabic and Byzantine coins have been recovered, as well as a fragment of scales, weights, silver bars and cowrie shells (Berga, 1988, p. 84). At Mežīte Hill-Fort a small excavation area (24 m²) produced seven 11 th and 13 th century coins, the mechanism of a set of collapsible scales and a fragment of a wrought silver bar (in addition to which an 11 th century coin and weights were found in the settlement site). This is somewhat surprising, since there are objectively greater opportunities for trading activity at centres located next to waterways. It is possible that trade had less importance at the centre represented by Beltes Hill-Fort and the settlement site, compared with Mežīte, while other functions, such as administrative power functions, were more important. 3. The small Puze Hill-Fort, established in the 12 th century, and burned down in the middle or second half of the13 th century, can be regarded as the residence of a local leader and his household. No settlement or village ever developed next to Puze Hill-Fort, so the village of Puze mentioned in 13 th century written sources may be thought to have been located about 1 km from here, at the later site of Puze Manor. 4. Judging from the ornaments found at Beltes and Mežīte Hill-Forts, and the material recovered from the burial sites at Mežīte, the Late Iron Age inhabitants of this site can be identified as belonging to a separate Finnic group possibly the Vends, mentioned in the Chronicle of Henry of Livonia. 5. The research to date indicates that the Couronian expansion, in the course of which they pushed into northern Kurzeme, began in the 11 th century and continued in the 12 th. This is clearly seen from the spread of characteristic Couronian flat cremation cemeteries in northern Kurzeme, in an area where inhumation had predominated in the 11 th and 12 th century. It is quite possible that Beltes Hill-Fort and settlement site, the material culture of which does not reflect Couronian influence, was destroyed in the 12 th century by the advancing Couronians. At this time a new centre was established at Veckuldīga, on the left bank of the River Venta, 6 km upstream from Beltes Hill-Fort. 6. Another previously disregarded phenomenon is observable in the material culture of the late phase of occupation, namely that less use was made of pottery vessels than in eastern Latvia. This is very apparent if we compare the quantity of potsherds found on various hillforts with the number of other artefacts recovered. At Puze Hill-Fort the number of potsherds was seven times greater (Vasks, 2004, p. 41), at Beltes Hill-Fort five times greater and at Mežīte Hill-Fort only twice as great as the number of other artefacts (Guščika & Vasks, 2010). Talsi Hill-Fort presents a very unusual case: the number of potsherds is only a quarter of that of other artefacts. The situation is completely different in the eastern part of Latvia. Thus, at Ķente Hill-Fort and settlement site the number of sherds was 12 times greater, at Asote and Lielkalni Hill-Forts 17 times greater (Šnore, 1961, p. 109, prilozenije II; Tora, 1996, p. 113), and at Dignāja Hill-Fort as much as 44 times greater than the number of other artefacts (Vasks, 2006 b, p. 111, 114). More examples could be found, but it is clear that the number of potsherds found at Iron Age residential sites in eastern Latvia is many times greater in relation to the number of other artefacts than it is in western Latvia. In other words, pottery vessels were used in greater quantities in eastern than in western Latvia. There are several possible explanations for such differences in material culture. One of these is that in the western part of Latvia vessels made of organic materials (wood, bark and leather) were much more widely used. The charred remains of turned wooden vessels and bark containers from Talsi Hill-Fort provide some indication of this (Karnups, 1938, 88). Of course, this does not mean that vessels of similar materials were not used in the eastern part of Latvia: it is simply a matter of the proportion of such vessels within the overall corpus of household vessels. 7. Significantly, in the Late Iron Age and at the beginning of the Middle Ages wheel-made pottery constitutes only 4 10 % of pottery on the hill-forts inhabited at this time in western Latvia, and is completely absent at Puze Hill-Fort. Only at Talsi Hill-Fort, inhabited right up to the 15 th century, does the proportion of wheelmade pottery reach 33 %. In eastern Latvia wheel-made pottery appeared in the second half of the 10 th and the 11 th century. Thus, at Oliņkalns, wheel-made pottery constituted % in the 11 th century (Mugurēvičs, 75

13 1977, p. 42). At Asote Hill-Fort it made up 78 % (Šnore 1961, p. 115) and at Jersika Hill-Fort as much as 90 % (Vilcāne, 2004, p. 54). At the Liv cemetery of Laukskola wheel-made vessels began to appear in graves in the second half of the 11 th century, and in the second half of the 12 th and 13 th century they constitute all the pottery (Zariņa, 2006, p. 312). Although the Late Iron Age and Early Medieval pottery from residential sites in western Latvia is so far quite scanty, it does seem that wheel-made pottery appeared later here, and handformed vessels remained in use longer. Perhaps the fact that in western Latvia wooden vessels were used to a greater degree in relation to pottery vessels can explain the slower introduction of the potter s wheel. Translated by Valdis Bērziņš Abbreviations APL Arheologu pētījumi Latvijā... gadā. Rīga, Literature Asaris J., Vasks A., Arheoloģiskie izrakumi Padures pilskalnā un apmetnē. In: APL , p Berga T. M., Бepгa Т. М., Монеты в ар - хео логических памятниках Латвии IX XII вв. Рига: Зинатне. Bērziņš V., Grasis N., Vasks A., Ziediņa E., Jauni 14 C datējumi arheoloģiskajiem pieminekļiem Rietumlatvijā. In: Latvijas Vēstures Institūta Žurnāls, 1, p Graudonis, J., Nocietinātās apmetnes Daugavas lejtecē. Rīga: Zinātne. Grigalavičienė E., Žalvario ir ankstyvasis ge ležies amžius Lietuvoje. Vilnius: Mokslo ir enciklopedijų leidykla. Guščika E., Vasks A., Pētījumi Mežītes arheoloģisko pieminekļu kompleksā un gadā. In: APL , p Karnups Ā., Izrakumi Talsu pilskalnā g. In: Se natne un Māksla, 2, p Mugurēvičs Ē., Oliņkalna un Lokstenes pilsnovadi gs. arheoloģiskie pieminekļi. Rīga: Zinātne. Rasiņš A., Tauriņa M., Pārskats par Latvijas PSR arheoloģiskajos izrakumos konstatētajām kultūraugu un nezāļu sēklām. In: Arheoloģija un etnogrāfija, 14, p Šnore E., Шнope Э. Д., Асотское го родище. In: Шноре Э. Д,, Зейд Т. Я. (ред). Материалы и исследования по археологии Латвийской ССР, II. Рига: Издательство Академии Наук Латвийской ССР. Tora A., Pētījumi Lielkanu pilskalnā. In: Zi nātniskās atskaites sesijas materiāli par arheologu un gada pētījumu rezultātiem. Rīga: Latvijas vēstures institūta apgāds, p Vasks A., Brikuļu nocietinātā apmetne. Lubāna zemiene vēlajā bronzas un dzelzs laikmetā (1000. g. pr. Kr g. pēc Kr.). Rīga: Preses Nams. Vasks A., Latvijas pilskalnu izpētes gaita. In: Latvijas Vēstures Institūta Žurnāls, 4, p Vasks A., Arheoloģiskie izrakumi Puzes pilskalnā. In: APL , p Vasks A., 2006 a. Arheoloģiskie izrakumi pie Ventas. In: APL , p Vasks A., 2006 b. Dignājas pilskalna apmetne un gada izrakumu gaismā. In: Pētījumi sēļu senatnē. Rakstu krājums. Latvijas Nacionālā Vēstures Muzeja Raksti Nr. 11. Arheoloģija, antropoloģija, etnogrāfija, folkloristika. Rīga: N.I.M.S., p Vasks A., Arheoloģiskie izrakumi Padures (Beltu) pilskalnā un gadā. In: APL , p Vasks A., Kalniņa L., Daugnora L., Beltu pilskalns. In: Arheoloģija un etnogrāfija, 25, in press. Vijups A., Puzes pilskalna depozīts. In: Ventspils muzeja raksti, III. Rīga, p Vilcāne A., Senā Jersika. Rīga: Latvijas vēstures institūta apgāds. Zariņa A., Salaspils Laukskolas kapulauks gadsimts. Rīga: Latvijas vēstures institūta apgāds. 76

14 Nauji duomenys apie Šiaurės Vakarų Latvijos piliakalnius Santrauka Nuo 2002-ųjų metų Latvijos universiteto Istorijos ir filosofijos fakulteto Archeologijos ir istorijos mokslų katedra organizavo kasinėjimus keliose Vakarų Latvijos vietovėse. Jų metu archeologiją studijuojantys studentai atlikdavo praktiką. Vyko trijų: Beltes, Mežīte ir Puze, piliakalnių archeologiniai kasinėjimai (1 pav.). Išsamiausi buvo Beltes (Padure) piliakalnio, kuris yra kairiajame Ventos upės krante, kasinėjimai. Visas piliakalnio aikštelės plotas m², kasinėjimai vyko 280 m² plote. Radinių pobūdis leidžia nustatyti dviejų apgyvendinimo etapų pėdsakus. Pirmasis priklauso ankstyvajai epochai: vėlyvajam žalvario ir ankstyvajam geležies amžiui, antrasis datuojamas vėlyvąja epocha: viduriniu ir vėlyvuoju geležies amžiumi. Vėlyvajame žalvario amžiuje piliakalnio apgyvendinta, lengviau prieinama dalis buvo apsaugota medinių stulpų eile, kurią buvus rodo išlikusių cm skersmens stulpaviečių eilė (3 pav.). Šalia jos rastas židinys, sukrautas iš akmenų, iš jo buvo paimtas medžio anglių mėginys. Jis datuotas m. prieš Kristų, tai yra žalvario amžiaus viduriu. Ankstyvojo apgyvendinimo metu buvusius pastatus žymi židiniai iš krautų akmenų ir stulpavietės, išsiskiriančios šviesios spalvos žemėje. Vis dėlto nepavyko nustatyti tikslesnio pastatų išsidėstymo. Keramika sudarė didžiąją dalį radinių, datuojamų ankstyvosios gyvenvietės laikotarpiu. Buvo rasta keletas iš titnago, ragų ir kaulo pagamintų įrankių. 60 % ankstyvosios keramikos buvo brūkšniuotu paviršiumi. Apie 11 % puodų šukių priklauso vadinamajai ankstyvajai grublėtajai keramikai. Naujas šio laikotarpio bruožas buvo tekstilinės keramikos, pirmą kartą surastos Vakarų Latvijoje, atsiradimas (apie 2 %). Sprendžiant pagal keramikos, rastos kai kuriuose židiniuose, tipus, pirmasis piliakalnio apgyvendinimo etapas baigėsi ankstyvojo geležies amžiaus pabaigoje. Kasinėjimų metu nebuvo surasta dirbinių, datuojamų I V amžiumi, todėl galima daryti prielaidą, kad šiuo laikotarpiu piliakalnis nebuvo apgyvendintas. VI ar VII amžiuje piliakalnyje prasidėjo antrasis apgyvendinimo etapas. Tai vėlyvasis apgyvendinimo laikotarpis. Intensyviausiai piliakalnis buvo apgyvendintas IX XII amžiuje. Dauguma rastų struktūrų ir radinių priklauso šiam laikotarpiui (5 pav.). Sprendžiant pagal radinius, X XII amžiuje Beltes piliakalnis tapo ekonominiu centru, kuriame buvo plėtojami amatai ir prekyba. Dirbinių chronologija ir radiokarbono datos nurodo vėlyvojo laikotarpio apgyvendinimo pabaigą, o kartu ir minėto ekonominio centro išnykimą. Dauguma radinių datuojami nuo vėlyvojo geležies amžiaus, tai yra IX XII amžiumi, ypač XI ir XII amžiaus pirmąja puse. Aštuonios radiokarbono datos patenka į laikotarpį tarp m. po Kristaus. O trys iš jų priklauso laikotarpiui tarp , tai yra, XII amžiui ir XIII amžiaus pradžiai. Remiantis šiais duomenimis, Beltes piliakalnis prarado ekonominio centro statusą XII amžiaus antroje pusėje arba XIII amžiaus pradžioje. Mežīte archeologinių objektų kompleksas, kurio centre stūkso piliakalnis, o jį supa maždaug 4 ha dydžio gyvenvietė. Į šį kompleksą įeina ir trys kapinynai ir kulto vieta Elkakalns. Nors žvalgomaisiais kasinėjimais buvo ištirtas tik 24 m² dydžio plotas iš piliakalnio aikštelę sudarančių m² ploto, archeologiniai radiniai buvo nepaprastai įdomūs. Iki 2 m storio kultūriniame sluoksnyje buvo rastos antžeminio tipo pastato su krautų akmenų židiniu liekanos. Pastato apačioje buvo aptikta 2 3 m dydžio, 80 cm gylio rūsio duobė (6 pav.). Nors kasinėjimai buvo vykdomi tik mažame piliakalnio plote, radiniai džiugino tiek gausa, tiek įvairove buvo rasta žalvarinių papuošalų, ginklų ir darbo įrankių (7 pav.) Daugelis radinių įrodo čia buvus prekybinę veiklą (septynios XI ir XIII amžiaus monetos, išardomų svarstyklių mechanizmas, taip pat svareliai ir nukirsta sidabro lydinio dalis). Elkakalns kulto vieta, trys kapinynai ir iš visų pusių piliakalnį juosianti didelė gyvenvietė, kurios kultūrinis sluoksnis siekia 1,5 m storio, visa tai įrodo, kad XI XIII amžiuje Mežīte buvo svarbus šio regiono centras, kuriame buvo gerai išvystyta prekyba. Greičiausiai piliakalnis ir pati gyvenvietė buvo sunaikinti kryžiuočių antpuolio metu XIII amžiuje. Neatmetama versija, kad šis piliakalnis yra pagrindinė castellaruta Lodgie, kuri minima 1234-ųjų metų dokumente, vieta ir gyventojų centro Ladze, paminėto 1253-iujų metų dokumente, vieta. Puze piliakalnis skiriasi nuo dviejų anksčiau aprašytų piliakalnių savo mažumu. Jo aikštelės plotas užima tik 600 m², kasinėjimai vyko tik 34 m² dydžio plote. Kultūriniame sluoksnyje, iki 30 cm gylyje, buvo surastos dviejų pastatų liekanos (8 pav.). Vienas šių pastatų, kurio matmenys 3,5 4 m, buvo aptiktas šalia piliakalnio šlaito. Manoma, kad jis priklausė išilgai piliakalnio aikštelės pastatytiems statiniams arba išorinei gynybinei sienai. Šio pastato vietoje ir šalia jo buvo aptikta daug sudegintų grūdų (paprastųjų kviečių ir rugių, kiek mažiau miežių; taip pat pašarinių pupų, žirnių, nedideli avižų kiekiai, tarp apanglėjusių augalų pėdsakų buvo rasta ir piktžolių). Kito pastato vieta aikštelėje išsiskyrė natūraliame žemės sluoksnyje susiformavusia tamsesne struktūra. Kasinėjimų metu nustatyta, kad jos plotis 1,6 m, o ilgis 2 m. Gali būti, kad čia buvo rūsys, panašus į Mežīte piliakalnio rūsius, ir kad pats pastatas buvo didesnis. Šioje vietoje buvo surasta 77

15 geležinių ir žalvarinių dirbinių. Šių radinių datavimas leidžia teigti, kad piliakalnyje buvo gyventa XII XIII amžiuje. Dauguma radinių buvo stipriai apdegę, kai kurie jų virtę beformiais susilydžiusio žalvario luitais. Iš to galima daryti prielaidą, kad piliakalnis buvo sunaikintas didelio gaisro. Šioje struktūroje buvo rasta ir dėžutė su žalvariniais moters papuošalais (9 pav.). Tarp jų buvo antkaklė, grandelių su kabučiais, pasaginių segių ir lankinė segė, taip pat žiedų. Iš sudegusio sluoksnio paėmus apanglėjusių grūdų mėginį, nustatytas gynybinių įtvirtinimų sunaikinimo laikotarpis m. po Kristaus, tai yra apie XIII amžiaus vidurį. Manoma, kad greičiausiai šis piliakalnis taip pat buvo sunaikintas puolančių kryžiuočių. Nors ir nedidelio masto, tokio pobūdžio piliakalnių tyrinėjimas suteikė keletą naujų įžvalgų apie Vakarų Latvijos priešistorę, kuri kol kas nėra plačiau tyrinėta: 1. Kaip parodė Beltes piliakalnio radiokarbono datos, įtvirtintos gyvenvietės arba piliakalniai Vakarų Latvijoje atsirado vėlyvojo žalvario amžiaus pradžioje, tai yra apie 2 tūkstantmečio pr. Kristų pabaigą maždaug tuo pačiu metu, kai jie pasirodė Rytų Latvijoje ir Lietuvoje. Radiniai Beltes piliakalnyje ir neįtvirtintoje Priednieki gyvenvietėje rodo, kad Šiaurės vakarų Latvija taip pat gali būti įtraukta ir į tekstilinės keramikos paplitimo teritoriją. 2. Vėlyvojo geležies amžiaus Beltes piliakalnis, o ypač Mežīte piliakalnis, gali būti priskirti prie pagrindinių centrų, turėjusių dideles gyvenvietes arba ankstyvuosius miestus, įsikūrusius šalia jų. Palyginus šalia svarbaus vandens kelio įsikūrusį Beltes piliakalnį su Mežīte piliakalniu, neturinčiu prieigos prie vandens, galima teigti, kad Mežīte yra daugiau duomenų, rodančių aktyvesnę prekybą, nei Beltes piliakalnyje. Tai gana stebinanti išvada, nes aktyvesnės prekybos galimybių turi centrai, įsikūrę šalia vandens kelių. Galima spėti, kad, lyginant su Mežīte, centre, kurį sudarė Beltes piliakalnis ir šalia esanti gyvenvietė, prekybai teko ne toks svarbus ekonominis vaidmuo. Šiuo atveju svarbesnės galėjo būti kitos funkcijos, pavyzdžiui, administracinė. 3. XII amžiuje įkurtas nedidelis Puze piliakalnis, sudegintas XIII amžiaus viduryje arba pabaigoje, gali būti laikomas vietinio vado arba jo šeimos gyvenamąja vieta. Aplink šį piliakalnį niekada nebuvo įsikūrusi jokia gyvenvietė ar kaimas, todėl daroma prielaida, kad XIII amžiaus rašytiniuose šaltiniuose minimas Puze kaimas galėjo būti įsikūręs už 1 km nuo šio piliakalnio, šalia vėliau įkurto Puze dvaro. 4. Sprendžiant pagal Beltes ir Mežīte piliakalniuose rastus papuošalus ir pagal Mežīte kapinynų radinius, gali būti daroma prielaida, kad šiose vietose įsikūrusios vėlyvojo geležies amžiaus gyvenvietės priklausė atskirai finų grupei tai galėjo būti ir Henriko iš Livonijos kronikoje minimi vendai. 5. Iki šiol vykdyti kasinėjimai rodo, kad kuršių ekspansija, kurios metu jie išsiplėtė į Šiaurės Kuržemę, prasidėjo XI amžiuje ir tęsėsi XII amžiuje. Tai akivaizdžiai įrodo kuršiams būdingi plokštiniai degintiniai kapinynai Šiaurės Kuržemėje, teritorijoje, kurioje XI ir XII amžiuje vyravo inhumacijos paprotys. Gali būti, kad Beltes piliakalnį ir gyvenvietę, kurių materialioji kultūra nebuvo patyrusi kuršių kultūros įtakos, XII amžiuje sunaikino puolantys kuršiai. 6. Vėlyvojo apgyvendinimo materialiojoje kultūroje atsispindi dar vienas bruožas, į kurį anksčiau nebuvo kreipta pakankamai dėmesio. Tuo laikotarpiu šiuose piliakalniuose buvo naudojama mažiau keramikos nei Rytų Latvijoje. Tai tampa akivaizdu, palyginus įvairiuose piliakalniuose rastų keramikos šukių kiekį su kitais rastais dirbiniais. Puzes piliakalnyje buvo rasta septynis kartus daugiau puodų šukių, Beltes piliakalnyje penkis kartus daugiau, o Mezite piliakalnyje tik du kartus daugiau, palyginti su kitais rastais dirbiniais. Talsi piliakalnis šiuo atžvilgiu labai neįprastas čia surastų puodų šukių kiekis sudaro tik ketvirtadalį visų kitų dirbinių. Rytinėje Latvijos dalyje yra visiškai kitaip. Šioje šalies dalyje buvo rasta kartus daugiau keramikos nei kitų dirbinių. Galima teigti, kad Rytų Latvijoje keraminių indų buvo naudota kur kas daugiau nei Vakarų Latvijoje. Tokius materialiosios kultūros skirtumus galima paaiškinti keliais būdais. Vienas jų kad Vakarų Latvijoje buvo daug dažniau naudojami iš organinių medžiagų pagaminti indai. Be abejo, tai nereiškia, kad iš panašių medžiagų pagaminti indai nebuvo naudojami rytinėje Latvijos dalyje. Kalbama tik apie keramikos dirbinių proporciją, palyginti su visais namų ūkyje naudotais indais. 7. Svarbu atkreipti dėmesį į faktą, kad vėlyvajame geležies amžiuje ir Viduramžių pradžioje žiestosios keramikos dirbiniai sudaro tik 4 10 % visų šiuo laiku apgyvendintų Vakarų Latvijos piliakalnių keramikos radinių, o Puze piliakalnyje jų apskritai nerasta. Rytinėje Latvijos dalyje žiestoji keramika atsirado X amžiaus antroje pusėje ir XI amžiuje. XI ir XII amžiuje žiestoji keramika sudarė % visų keramikos dirbinių rytinės Latvijos piliakalniuose. Nors gyvenamosiose Vakarų Latvijos vietovėse rasta nedaug vėlyvojo geležies amžiaus ir ankstyvųjų Viduramžių keramikos, atrodo, kad žiestoji keramika čia atsirado vėliau, o lipdytinė keramika buvo naudojama ilgiau. Galbūt Vakarų Latvijoje buvo naudojama daugiau medinių indų, palyginti su keraminiais indais, dėl to žiedžiamasis ratas tik vėliau prigijo šioje vietovėje. Iš anglų kalbos vertė Auksė Pūraitė-Kuokštienė Įteikta 2011 m. vasario mėn. 78